首页 | 本学科首页   官方微博 | 高级检索  
     

基于顺序形态滤波的外运动小目标检测
引用本文:叶斌,彭嘉雄,卢汉清.基于顺序形态滤波的外运动小目标检测[J].数据采集与处理,2001,16(3):315-319.
作者姓名:叶斌  彭嘉雄  卢汉清
作者单位:华中理工大学图像识别与人工智能研究所,
基金项目:国家重点实验室开放课题基金资助项目.
摘    要:针对约外图像中运动弱小目标的检测问题,本文提出了基于顺序形态滤波的小目标检测方法,并给出了具体算法。目标检测分两步进行,首先利用顺序形态滤波抑制背景,检测出候选目标的位置,并对候选目标进行区域分割,最后利用序列图像中目标运动的连续性和轨迹的一致性筛选出真正的目标,作者通过实验比较了该方法与传统高通滤波方法在抗噪声性能,背景抑制性能以及抑制虚警目标性能的差异,实验结果表明,顺序形态滤波法在这三个方面都优于高通滤波法,它能够快速、可靠检测出低信噪比的运动小目标。

关 键 词:小目标检测  高通滤波  顺序形态滤波  图像流  红外图像  序列图像  模式识别
文章编号:1004-9037(2001)03-0315-05
修稿时间:2000年9月20日

Moving Small Target Detection Based on Order Morphology Filtering in Infrared Image Sequences
Ye Bin Peng Jiaxiong Institute for Pattern Recognition and Artificial Inte lligence,Huazhong University of Science and Technology Wuhan ,P.R.Chi na Lu Hanqing The National Laboratory of Pattern Recognition,The Institute of Automation,Chines.Moving Small Target Detection Based on Order Morphology Filtering in Infrared Image Sequences[J].Journal of Data Acquisition & Processing,2001,16(3):315-319.
Authors:Ye Bin Peng Jiaxiong Institute for Pattern Recognition and Artificial Inte lligence  Huazhong University of Science and Technology Wuhan  PRChi na Lu Hanqing The National Laboratory of Pattern Recognition  The Institute of Automation  Chines
Affiliation:Ye Bin Peng Jiaxiong Institute for Pattern Recognition and Artificial Inte lligence,Huazhong University of Science and Technology Wuhan 430074,P.R.Chi na Lu Hanqing The National Laboratory of Pattern Recognition,The Institute of Automation,Chines
Abstract:To resolve the detection of moving small target in infr ared image sequences, a new small target detection method based on order morphol ogy filtering is presented, and the detailed algorithm is given. Two steps of ta rget detection are adopted. First suppressing background and determining the loc ation of the candidate targets by using order morphology filtering, and segmenti ng the regions where there are possible targets are performed. Finally the true targets are screened according to the principle of moving continuity and traject ory consistency of moving target in the image sequences. Experiments show th at the method is more effective in three performances of anti-noise, suppressin g background and suppressing the number of false targets than that of high pass filterin g method. It can effectively and reliably detect the moving small target with lo w SNR.
Keywords:small target detection  high pass filtering  o rder morphology filtering  image flow
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号